Early prediction of pancreatic cancer through detection of methylated circulating DNA in blood.

Abstract

e16353 Background: Pancreatic cancer represents a significant oncological challenge, contributing markedly to cancer-associated mortalities. This is attributed to the asymptomatic nature of early-stage disease, its inherently aggressive phenotype, and a notably suboptimal response to existing systemic therapies. Hence, the formulation of efficacious early diagnostic modalities for pancreatic cancer amenable to surgical resection is imperative to augment patient prognoses and enhance survival outcomes. Methods: The Infinium MethylationEPIC v2.0 BeadChip platform facilitated the examination of methylation profiles within plasma samples, while the HumanMethylation450 BeadChip array was employed for analogous analyses in corresponding tissue specimens from Taiwanese pancreatic cancer cohorts, aiding in gene selection. Subsequent methylation status verification was conducted via quantitative Methylation-Specific PCR (qMSP) on both genomic DNA and circulating cell-free DNA samples. For external methylation confirmation within a Western demographic, data from The Cancer Genome Atlas (TCGA) Portal were utilized, with further validation conducted using plasma samples from U.S. pancreatic cancer patients. Total 105 plasma from pancreatic cancer patients and healthy subjects were analyzed to calculate the candidate the accuracy, sensitivity, specificity of biomarkers for predict the pancreatic cancer. Results: The test exhibited an accuracy of 90.4%, sensitivity of 92.3%, and specificity of 90.1% in the detection of pancreatic cancer. Detection levels increased from 0% in healthy individuals to 87.5% in Stage I and 100% in Stages II, III, and IV. The CA19-9 marker displayed limited specificity with a high rate of false positives. Moreover, the test dynamically evaluated tumor burden by monitoring circulating methylated ZFP30 and FBXL7 in the plasma of breast cancer patients post-treatment. Conclusions: The analysis of plasma for methylated ZFP30 and FBXL7 provides an innovative and accurate noninvasive method that outperforms the existing blood biomarker CA19-9 for the prediction and ongoing assessment of tumor burden following treatment in pancreatic cancer.